You are tasked with loading data from an external stage containing gzipped CSV files into a Snowflake table. The CSV files have a header row that you want to skip during the load process. Additionally

Question

Options

• AFILE_FORMAT = (TYPE = CSV, SKIP_HEADER = 1 , COMPRESSION = 'GZIP')
• BTRANSFORM = (FULL_NAME = FIRST_NAME II ' ' II LAST_NAME)
• CON ERROR = 'SKIP FILE'
• DTRANSFORM AS SELECT $1, $2, $3 II ' II $4 FROM @mystage/data.csv
• EFILE FORMAT = (TYPE = CSV, SKIP_HEADER = 1, COMPRESSION = 'GZIP'), TRANSFORM

• A
  79% (22)
• B
  7% (2)
• C
  11% (3)
• E
  4% (1)

Explanation

Options A and D together cover both requirements: A correctly specifies FILE_FORMAT with TYPE = CSV, SKIP_HEADER = 1 to drop the header row, and COMPRESSION = 'GZIP' for the compressed files - all in valid Snowflake syntax. D uses Snowflake's inline SELECT transformation pattern, referencing positional columns ($3, $4) and concatenating them with || (shown as II in the question), which is exactly how COPY INTO handles column-level transforms from a stage.

B is wrong because Snowflake has no TRANSFORM = (key = expression) option - that syntax doesn't exist; transformations must be a SELECT subquery in the FROM clause. C (ON ERROR = 'SKIP FILE') is a valid COPY INTO clause for error handling, but it does nothing for header skipping or column concatenation - it's a distractor that sounds plausible. E fails on two counts: FILE FORMAT (with a space) is a syntax error - Snowflake requires FILE_FORMAT (underscore) - and the trailing TRANSFORM keyword without a SELECT expression is not valid syntax.

Memory tip: "Format in FILE_FORMAT, transform in FROM (SELECT)" - Snowflake COPY INTO keeps these concerns separate; there is no magic TRANSFORM = shorthand.

No community discussion yet for this question.